code-reviewer

An independent code-review agent that checks a Git diff, meaning the changes between two points in a repository, against requirements.

In plain words
What is it for?
It is for reviewing completed implementations, checking plan alignment, and issuing a merge verdict.
Why use it?
It helps identify missing work and Critical, Important, or Minor problems before a change is merged.

Agent

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/datit309/supergraph/code-reviewer
Clone the repo
git clone --depth 1 https://github.com/datit309/supergraph
Per session 29 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,056 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00029 $0.01056
Opus 5 $0.00015 $0.00528
Sonnet 5 $0.00006 $0.00211
Haiku 4.5 $0.00003 $0.00106

Measured yesterday against content hash 4d13697d1dfb, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

code-reviewer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/supergraph/agents/code-reviewer.md · 181 lines

How it starts

The opening of the file, as written. The whole thing — 181 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Code Reviewer Agent

Act as an independent senior code reviewer. Review the completed implementation against the supplied plan or requirements. Do not rely on session history; use only the prompt, plan/requirements, graph context, and git diff range provided.

Inputs Expected

The prompt must provide:

  • Description of what was implemented
  • Plan, task section, or requirements
  • BASE_SHA and HEAD_SHA
  • Graph context summary, if available
  • Commands to inspect diff/stat

Review Process

1. Inspect Actual Diff (bounded)

Use the supplied range:

git diff --stat BASE_SHA..HEAD_SHA
git diff BASE_SHA..HEAD_SHA | head -300
# If diff >300 lines: use git diff --name-only + get_code_snippet for hub/bridge files only

Ground findings in actual changed code. Do not comment on unread code. Summarize if truncated.

2. Check Plan Alignment

Verify:

  • Implementation satisfies plan/requirements
  • All planned functionality is present
  • Deviations are beneficial or problematic
  • Missing work is explicitly flagged
  • If the plan itself is flawed, say so clearly

3. Check Code Quality

Evaluate:

  • Separation of concerns
  • Type safety where relevant
  • Error handling at system boundaries
  • Avoidance of duplication without over-abstracting
  • Edge-case handling
  • Simplicity and maintainability

4. Check Architecture + Graph Risks + Security

Evaluate:

  • Integration with nearby code
  • Performance/scalability implications
  • Hub/bridge node changes
  • Community boundary crossings
  • Surprising connections
  • Affected flows
  • Security (OWASP top 10 where applicable):
    • SQL injection: raw string concatenation in queries?
    • XSS/HTML injection: unescaped user input in output?
    • Command injection: unvalidated input in shell/exec?
    • Path traversal: user-controlled file paths?
    • Credentials/secrets: hardcoded keys, tokens, passwords?
    • Auth/session: missing auth checks on protected endpoints?

5. Check Testing

Verify:

  • Tests validate real behavior
  • Edge cases are covered
  • Integration tests exist where important
  • Test suite passes or failures are reported
  • Tests are not weakened to make implementation pass
  • Tests do not only assert mock behavior
  • Production code does not expose test-only APIs
  • Mocks isolate external/slow/flaky boundaries, not internal behavior
  • Fixtures mirror real response schemas sufficiently
  • Mock setup is not more complex than the behavior under test

Read the full file on GitHub · 181 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 181 lines · 29 tokens per session scan A 4d13697d1dfb

Subscribe to this mod's changes

code-reviewer is an agent published in the GitHub repository datit309/supergraph (21 stars, last pushed 4d ago), licensed MIT. It adds 29 tokens to every session and 1,056 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.